Qwen3.5-4B
MultimodalQwen3.5-4B is a 4 billion parameter vision-language model using Gated DeltaNet hybrid architecture with a 3:1 ratio of linear attention to full softmax attention. It supports 262K native context length and delivers strong performance for its size across knowledge, reasoning, coding, and multilingual tasks.
Key Specifications
Parameters
4.0B
Context
-
Release Date
March 2, 2026
Average Score
64.4%
Timeline
Key dates in the model's history
Announcement
March 2, 2026
Last Update
September 10, 2026
Today
September 19, 2026
Technical Specifications
Parameters
4.0B
Training Tokens
-
Knowledge Cutoff
-
Family
-
Capabilities
MultimodalZeroEval
Benchmark Results
Model performance metrics across various tests and benchmarks
Reasoning
Logical reasoning and analysis
GPQA
Thinking mode • Self-reported
Other Tests
Specialized benchmarks
AA-LCR
Thinking mode • Self-reported
BFCL-V4
Thinking mode • Self-reported
C-Eval
Thinking mode • Self-reported
DeepPlanning
Thinking mode • Self-reported
Global PIQA
Thinking mode • Self-reported
HMMT 2025
February 2025 • Self-reported
HMMT25
November 2025 • Self-reported
IFBench
Thinking mode • Self-reported
IFEval
Thinking mode • Self-reported
Include
Thinking mode • Self-reported
LiveCodeBench v6
Thinking mode • Self-reported
LongBench v2
Thinking mode • Self-reported
MAXIFE
Thinking mode • Self-reported
MMLU-Pro
Thinking mode • Self-reported
MMLU-ProX
Thinking mode • Self-reported
MMLU-Redux
Thinking mode • Self-reported
MMMLU
Thinking mode • Self-reported
Multi-Challenge
Thinking mode • Self-reported
NOVA-63
Thinking mode • Self-reported
PolyMATH
Thinking mode • Self-reported
SuperGPQA
Thinking mode • Self-reported
t2-bench
Thinking mode • Self-reported
VITA-Bench
Thinking mode • Self-reported
WMT24++
Thinking mode • Self-reported
License & Metadata
License
apache_2_0
Announcement Date
March 2, 2026
Last Updated
September 10, 2026
Similar Models
All ModelsQwen3 VL 4B Thinking
Alibaba
MM4.0B
Best score:0.8 (MMLU)
Released:Sep 2025
Price:$0.10/1M tokens
Qwen3 VL 4B Instruct
Alibaba
MM4.0B
Best score:0.8 (MMLU)
Released:Sep 2025
Price:$0.10/1M tokens
Qwen2.5 VL 7B Instruct
Alibaba
MM8.3B
Released:Jan 2025
Qwen3.5-2B
Alibaba
MM2.0B
Best score:0.5 (GPQA)
Released:Mar 2026
Qwen2.5-Omni-7B
Alibaba
MM7.0B
Best score:0.8 (HumanEval)
Released:Mar 2025
Qwen3 VL 32B Thinking
Alibaba
MM33.0B
Best score:0.9 (MMLU)
Released:Sep 2025
Qwen2.5 VL 32B Instruct
Alibaba
MM33.5B
Best score:0.9 (HumanEval)
Released:Feb 2025
Qwen3.5 9B
Alibaba
9.0B
Best score:0.8 (GPQA)
Released:Mar 2026
Recommendations are based on similarity of characteristics: developer organization, multimodality, parameter size, and benchmark performance. Choose a model to compare or go to the full catalog to browse all available AI models.